Adhesive interactions between tumor cells and the extracellular matrices (ECMs) of their host environment are generally believed to be significant factors in metastasis, although the evidence for this is largely circumstantial. The aim of this project is to address the question directly: To what extent do specific cell:ECM interactions influence the metastatic behavior of tumor cells? To accomplish this, tumor cell lines will be generated (by in vitro selection) which differ in specific adhesive capacities, but which are in all other respects as closely identical as possible. These substratum-specific adhesion variants will be selected from established malignant tumor cell lines such as the B16 melanoma, RAW117 lymphosarcoma, and rat 13762 mammary adenocarcinoma. The variants will then be compared with each other and with the parental cells in terms of in vivo malignancy characteristics. Long-term goals include the characterization of these variants in biochemical terms, in order to identify specific cellular factors involved in adhesive cell interactions with ECMs. This approach will permit the direct characterization of individual contributions of specific cell:ECM interactions to the metastatic potential of tumor cells. This in turn may stimulate the development of new approaches to the management of metastatic disease. (A)